Vezina-worthy: Using Advanced Data to Determine the Rightful Winner

Biases arose and became unwritten rules (only playoff goalies qualify). The best goaltender often didn’t win (Jim Carey beat Dominik Hasek). Narratives began to matter more than numbers (check out Martin Brodeur’s save percentage rank in 2003 and 2004). 

In short, mistakes were made.

The Goaltending Enlightenment (2015-present) has seen an explosion of new ways to measure goaltending success, from strength-specific save percentages, to shot location heatmaps, to expected goals models featuring delta save percentage and goals saved above average. The affect of those statistics on Vezina voting was surprising.

Surprise! They made no difference whatsoever. As it turns out, NHL general managers have found a method that works for them, and gosh darn it, they’re sticking with it. 

So. 

If you’re looking to predict who will win the Vezina trophy in a given season, I’ve already published the guide you need – check it out here. It’s five years old, and hasn’t required updating till now. I’m hopeful the changing nature of the position and the evolution of the game will force general managers to consider different voting criteria. Maybe this will be the year all the GMs subscribe to InGoal Magazine and see a whole new world of possibility. Fingers crossed.

Regardless, I’m not here to make predictions. Today, we’re going on a journey to determine, using the best data available, which goalies outperformed their peers and truly deserve to be in the Vezina conversation. Forget about who will win – we’re going to find out who should win.